53

Around a year ago, I reviewed a paper for a journal. The originality of the paper was questionable, and content and presentation were severely lacking throughout. Consequently, both another reviewer and I recommended rejection and gave detailed explanations with which the associate editor agreed. The decision, including the review reports, were sent to the authors.

Now, a colleague of mine just told me of a review request from another journal. Briefly summarizing the content of the paper, I realized that this sounded quite familiar. Expressing my concerns to my colleague, we compared the authors and the papers, and realized that this was exactly the same paper that was reviewed and rejected earlier; nothing has been changed (apart from some journal style-specific things).

How should we proceed?

cpit
  • 588
  • 1
  • 5
  • 19
  • 79
    Probably would be inappropriate to influence your colleague's review of the paper... evidently the authors did not agree with your appraisal, rightly or wrongly, and resubmitted elsewhere. This is allowed. If the paper has serious problems beyond matters-of-opinion, your colleague will see that without prompting, and say so in the review. Done. No issue. – paul garrett Mar 08 '16 at 15:36
  • 1
    There was a case where something like this happened to me. But in this case the second journal sent it to me again. (I guess I was the natural referee.) This made it easy for me to do the second review... – GEdgar Mar 08 '16 at 16:07
  • 25
    As far as I know, there's nothing ethically wrong with submitting a rejected paper without changes to a different journal. It's not a smart thing to do in most cases, but it's not unethical. So there's nothing for you to do but step back and let your colleague review the paper on its merits (or lack thereof). –  Mar 08 '16 at 16:11
  • 17
    @user3825755, first of all, are you really allowed to talk about the papers you had reviewed in the past? Or even more critically, is your colleague allowed to talk about the paper that he/she is reviewing? In most of the journals I know or know of, I am not allowed to do so at any cost, even after reviewing it - this is built-in in the single-blind policy, i.e., the author(s) is not allowed to know who the reviewers are or were. If the two journals in your question have such a policy, then you and your colleagues have clearly violated it. – John Mar 08 '16 at 18:45
  • 4
    If your colleague is talking about the paper that he\she is currently reviewing, it is possible that the author of the paper may know who the reviewer was. Such a violation does have consequences that I have heard of - the journal banned the particular reviewer from reviewing or submitting the papers to this journal. In your case, the risk is double that since there are two journals involved. The other aspect is, as @paulgarrett mentioned, you shouldn't influence your colleague's decision. You could very well be a 'competitor' or 'enemy' of the author - one of the reasons for the above policy! – John Mar 08 '16 at 18:50
  • 4
    @John, indeed, I think that (e.g., in the U.S.) people submitting manuscripts to journals (e.g., in math) have a reasonable expectation of confidentiality, that would preclude any such discussions. Similarly with grant proposals and such. A poor quality submission is not entitled to less... or, as these things usually go, it's a slippery slope: violation of rules is rationalized by declaring people outside the protections ... of the rules. – paul garrett Mar 08 '16 at 18:55
  • FYI: What percentage of papers submitted to a conference or journal have been previously rejected in the same or another venue? – Franck Dernoncourt Mar 08 '16 at 21:59
  • 3
    @paulgarrett: even aside from a slippery slope argument, you can say that the value of the confidentiality is considered higher than the unfortunate cost in duplicated effort. Therefore, seeking to avoid a re-review of an identical submission isn't a legitimate reason to breach the confidentiality (although the questioner presumably started out in disagreement with this principle or they wouldn't be seeking to do something about it). – Steve Jessop Mar 09 '16 at 00:49
  • It seems that many of the answers disregard the reason for the rejection, namely the originality of the paper being in question. Could you clarify how much in question? I would say that if the material was shown by the reviewer comments to be already known then just resubmitting is unethical, but if is is just a bit unclear precisely how much "new" is in the paper, it is less clear. – Tobias Kildetoft Mar 09 '16 at 09:14
  • 1
    @TobiasKildetoft which ethical principle would submitting to a new journal without revision be violating, as I just don't see it? – Scott Seidman Mar 09 '16 at 11:29
  • 4
    @ScottSeidman When submitting a paper, you are claiming that the work presented is new (or if it is not, you obviously need to point out in the paper who did it prior to you). If it is pointed out to you by reviewers that the work is not new, then it is unethical to attempt to pass it off as new to another journal. – Tobias Kildetoft Mar 09 '16 at 11:42
  • @TobiasKildetoft yes, the work is new, as in unpublished. You are not swearing, or even suggesting, that the work has not been submitted before. You are only promising it is unemcumbered, and that the journal has legal right to publish it until such time as they waive that right by turning it down. There are no ethical considerations here. It's not the most productive use of anyone's time, but not ethically questionable. – Scott Seidman Mar 09 '16 at 11:53
  • 4
    @ScottSeidman No, if you are aware that the results are in fact not new (as in, have been published by someone else before you), then trying to publish those results as if you were unaware of this fact is unethical. – Tobias Kildetoft Mar 09 '16 at 11:56
  • @TobiasKildetoft, who is asserting that the work was previously published? I haven't seen anyone say that. – Scott Seidman Mar 09 '16 at 12:08
  • 4
    @ScottSeidman That is why I asked for a clarification of this. The originality being in question can either mean that the work is not original, or that it is only slightly original. The difference is pretty big. – Tobias Kildetoft Mar 09 '16 at 12:09
  • @ScottSeidman: While it is true that authors can know and declare that a manuscript is not under consideration for another journal, the corresponding question for reviewers is more tricky: I review for a journal that asks for reporting of manuscripts that I know are under consideration somewhere else. As a reviewer, I typically do not know when a manuscript stops being under consideration - so I'd have to report that I met the manuscript before but do not know whether it is still under consideration. – cbeleites unhappy with SX Mar 10 '16 at 11:53
  • Here everyone knows the paper was rejected, as the reviewer has seen the decision letter @cbeleites – Scott Seidman Mar 10 '16 at 11:56
  • (getting slightly off topic) @ScottSeidman: Yes, I read that the OP knows of the decision - which is something I've never encountered with the journals I review for. I conclude that the details of ethical conduct and confidentiality seem to vary considerably. – cbeleites unhappy with SX Mar 10 '16 at 12:00

4 Answers4

91

If you were asked to re-review an "unchanged" manuscript there are a number of things you can do (e.g., Asked again to review a paper, when the authors don't wish to modify it). The issue is that this is not the case. You are no longer part of the review process.

The first thing you should do is STOP. The behavior you have engaged in so far has been completely unethical and a clear violation of every reviewer agreement I have ever seen.

  • Your colleague should never have told you about the paper under review.
  • You should not have mention that you reviewed a similar paper in the past.
  • Neither of you should have mentioned authors or the title.
  • The actual manuscripts should never have been shared and/or compared

To a lesser extent, it is not even clear why you still have your copy of the manuscript.

The Committee on Publication Ethics (COPE) provides Ethical Guidelines for Peer Reviewers which can be thought of as best practice. These include:

  • respect the confidentiality of peer review and not reveal any details of a manuscript or its review, during or after the peer-review process, beyond those that are released by the journal

  • not involve anyone else in the review of a manuscript, including junior researchers they are mentoring, without first obtaining permission from the journal; the names of any individuals who have helped them with the review should be included with the returned review so that they are associated with the manuscript in the journal’s records and can also receive due credit for their efforts.

  • keep all manuscript and review details confidential.

Community
  • 1
StrongBad
  • 104,237
  • 30
  • 262
  • 484
  • 4
    I agree with your answer, thanks for the plain-spoken reply –  Mar 08 '16 at 19:34
  • 22
    This sounds like a step-by-step instruction to push unsound papers through the review process by retrying just for long enough until the authors are lucky to encounter a set of reviewers who happen not to notice the flaws. – O. R. Mapper Mar 08 '16 at 20:25
  • 35
    @O.R.Mapper there are lots of unethical ways to get unsound papers published and from my understanding the peer-review process was not designed to protect against unethical authors. That said, in my opinion, disagreeing with reviewers is not unethical. – StrongBad Mar 08 '16 at 20:34
  • 6
    @StrongBad My understanding is that reviewing is kept anonymous to protect the reviewer from retaliation by the author, therefore allowing negative reviews if appropriate. Your points sound completely over the top to me. Also, there are obvious cases where this must not be followed, such as when a new researcher is learning how to do a referee report. – Jessica B Mar 10 '16 at 07:24
  • 3
    I just want to point out that not all journals do have an absolute confidentiality requirement: For example, the Analyst's reviewer guidelines say: "except in special cases, where specific scientific advice may be sought". (The editor has to be told that and who was consulted, though). So in case the OP is one of the "natural reviewers" and in case the journal in question has a similar policy, it is OK for the colleague to ask the OP his expert opinion. – cbeleites unhappy with SX Mar 10 '16 at 11:24
  • 1
    The Analyst's policies also explicitly ask "To inform the editor of any similarity between the submitted manuscript and another either published or under consideration by another journal". The consideration part here is IMHO tricky: as a reviewer I'm typically not informed on the decision of the editor - so once I got a manuscript, there is typically no way to know that a manuscript is not under consideration any more - which would create an ongoing duty to report that I've met the manuscript before. Taking these points together, sorry, I downvote as the answer is too absolute. – cbeleites unhappy with SX Mar 10 '16 at 11:36
  • 3
    @JessicaB there is a big difference between anonymous and confidential. This answer has nothing to do with the anonymity of the reviewer, it is about the confidentiality of the manuscript. As for new researchers, you need to ask the editor if you can consult someone. They almost always say yes. – StrongBad Mar 10 '16 at 13:16
  • 1
    @StrongBad I guess I don't tend to see that as a big issue since confidentiality of the manuscript makes no sense when the paper is already freely available on the internet. Usually I could legitimately read the paper without being asked to referee it. Indeed, a couple of times I had already done so. – Jessica B Mar 10 '16 at 15:09
  • 5
    it is not even clear why you still have your copy of the manuscript - is there any journal asking that reviewers should delete / destroy their copy of the manuscript after review? – silvado Mar 10 '16 at 19:56
  • @silvado I have never been told to destroy/delete the manuscript, but why would you keep it? COPE says that post review reviewers should "continue to keep details of the manuscript and its review confidential". – StrongBad Mar 10 '16 at 20:14
  • 2
    In that case, there's no need to question that the OP still has their copy of the manuscript in your answer. – silvado Mar 10 '16 at 21:05
  • @silvado: the policy I linked above does ask for destruction of the manuscript. I file the review I wrote, though - in case questions come up I can refer to my record. – cbeleites unhappy with SX Mar 10 '16 at 22:15
  • 1
    You mean the COPE policy? I can't find this statement there. Getting rid of something is not the same as keeping it confidential. I see the latter, but not the former. – silvado Mar 11 '16 at 10:14
  • 2
    Even in the contrary, the COPE policy formulates several post-review expectations for which it seems to be helpful to keep the manuscript on file. – silvado Mar 11 '16 at 10:17
  • Agree with everything except "it is not even clear why you still have your copy of the manuscript". There is nothing unethical about keeping a copy of the manuscript on a secure computer, as long as you keep everything confidential. Early access to new knowledge is one of the only personal benefits of donating your labor for free! Building on and referring back to that knowledge is perfectly fine as long as you don't publish anything before that manuscript comes out. – WetlabStudent Apr 12 '19 at 01:02
44

Between the two reviewers, you've created quite a nasty situation. First -- the author did NOTHING wrong.

Second, you and your colleague have done something very wrong. The fact that you know nothing substantive has changed means you were essentially handed the manuscript, which is very bad behavior on both your parts.

My recommendation is that the new reviewer should probably contact the editor that sent him the manuscript and say simply "for reasons I choose not to discuss, I suddenly find myself in conflict, and can't provide a review", delete the paper, and never discuss it again. Your colleague is not in a situation where he should try to provide a fair review, as he's obviously poisoned.

Scott Seidman
  • 31,120
  • 4
  • 52
  • 121
  • 4
    Exactly. All accidental, and perhaps with good intentions, but, nevertheless, ... – paul garrett Mar 08 '16 at 20:57
  • 8
    "Your colleague is not in a situation where he should try to provide a fair review, as he's obviously poisoned." - I suppose this hinges on whether the review is supposed to be a test of the reviewer's knowledge, and nothing else (so using information the reviewer did not find on their own could be called "unfair" and thus spoils the result), or a test of the manuscript's quality (so chances are, the result becomes more accurate if the reviewer receives additional information). – O. R. Mapper Mar 08 '16 at 21:12
  • 1
    Other than that, I find your claim that "the author did NOTHING wrong" interesting, in the light of answers I received on a somewhat related question of mine. – O. R. Mapper Mar 08 '16 at 21:15
  • 7
    @O.R.Mapper -- First, "wrong" is very different from "failed to take advantage of an opportunity to make something clearer". The author has an unencumbered manuscript, and it is certainly within his rights to submit it wherever he wants without having former reviewers influencing new reviewers for no reason. Second, I really don't see the parallel in your somewhat related question. The issues at hand here are reviewer misconduct, not the actions of the author. Had you submitted to a new journal, nobody would have questioned your right to do so without revision. – Scott Seidman Mar 08 '16 at 22:16
  • 2
    @ScottSeidman: It's not "failed to take advantage of an opportunity to make something clearer", it's "failed to make something as clear as it should have been". Which seems quite like a way of doing something "wrong". Then, the manuscript is not exactly "unencumbered" if possible flaws have already been found therein. The issue at hand seems to be authors-resubmit-until-manuscript-slips-through, and unless the primary reason for rejection is failing to address the journal topic, I do not see how this strategy can become ethical by trying to conceal it and switching target journals. ... – O. R. Mapper Mar 08 '16 at 23:08
  • 1
    ... The parallel that I see to my question is that the bottom line of most answers I received was that authors should almost never completely ignore any reviewer comments. I do not necessarily agree with this, but as an author, I also do not think exchanging reviewers until I find some that review to my liking is the way to go. If this "reviewing to my liking" happens because of different opinions, I consider the practice borderline, but once it gets as bad as expecting that the reviewers do not find a flaw that others have found before, I consider it downright frauduluent. – O. R. Mapper Mar 08 '16 at 23:11
  • 1
    @O.R.Mapper - we have NO WAY of knowing if the reviewer is right or wrong here -- and no way of knowing the basis of the criticism, or the standards of the two journals. The author, though, is entitled to a review based upon the merits of the ms, and not one based upon the collusion of members of his community. There is nothing unethical about submitting a manuscript that has been reviewed and rejected (though it may not be the most effective use of anyone's time or effort). There IS something unethical about agreeing to conduct a review on a confidential ms and sharing the ms. – Scott Seidman Mar 08 '16 at 23:17
  • 6
    @O.R.Mapper The review process is of course supposed to be a test of the manuscript's quality, but individual reviews should be independent of one another. By having two reviewers confer over the manuscript, you end up with correlated reviews, which is surely undesirable. – Will Vousden Mar 09 '16 at 12:56
  • 5
    @O.R.Mapper "Should" is being used in difference sense though. "most answers I received was that authors should almost never completely ignore any reviewer comments." -- "should" there means "unwise", not "unethical". It was probably unwise of the author to keep resubmitting the paper without taking prior reviews into consideration. That doesn't make it unethical. The author isn't the one asking the question, therefore what is wise or unwise for the author to do is irrelevant. – LindaJeanne Mar 09 '16 at 15:26
  • You only get a problem of correlation if there are two reviews to correlate. In this scenario, that objection doesn't hold. (separate journals, entirely) Whether or not the conversation unduly influenced the more recent reviewer can be argued, but not whether that constitutes any sort of violation of independence of reviewers. – The Nate Mar 10 '16 at 19:49
7

The goal of the review process is to fairly and accurately evaluate the merits of the submitted manuscript, while making sure no one gains an unfair advantage through knowledge of the manuscript before it is available publicly.

I see no issue about unfair advantage here, since you and your colleague were both already in possession of the same manuscript. That being said, ethical boundaries are very field dependent, and the culture in your field may be different. For example, I have reviewed a number of papers and have never been explicitly asked to keep submitted manuscripts confidential (though it is generally understood that I should).

As for your colleague's responsibility of evaluating the paper, there is some unfortunate tension between the goals of fairness and accuracy, and you need to make a judgement based on the specifics of the situation. But here are the main points I think are important:

  1. It is unreasonable to expect every reviewer to understand every tool used in a submitted paper. Discussion of papers (which both parties already have access to) is to be encouraged (though the fact that one is reviewing the paper being discussed should perhaps be kept confidential, depending on the situation). From this point of view, I would consider it unethical not to let your colleague know about a serious logical error in a paper they are reviewing.

  2. On the other hand, your colleague should form their own critical opinion about the paper. Their knowledge that the paper was previously rejected and then resubmitted without any revision has (probably) already biased them against it.

So, if there is a serious issue in the paper which absolutely has to be pointed out, then by all means do so. Otherwise, I would do as the other answers suggest and let your colleague form their own opinion about it.

The fact that the authors have not addressed your original reasons for rejection is unsettling and could be a reflection of unethical behavior on their part, but without more information, we (and perhaps you) cannot know for sure.

Scott Seidman
  • 31,120
  • 4
  • 52
  • 121
John Pardon
  • 167
  • 2
  • 6
  • 1
    Your field has some very liberal ethics compared to mine. In my field, if I want to discuss a paper with a colleague, I need to mm pass it by the editor. – StrongBad Mar 09 '16 at 10:33
  • 2
    One difference in math might be that the paper is usually already publicly available on the arXiv. (For example, the suggestion that one might discuss a paper as long as one hides that one is refereeing it.) – Tom Church Mar 11 '16 at 00:20
4

"One Man's Ceiling Is Another Man's Floor". A paper unsuitable for one kind of journal might be suited to another. While non unethical, submitting the same paper suggest a lack of critical thinking by, at least, the submitting author.

[EDIT] I do undertand, though, that, depending on the domain, authors may consider their paper could be resubmitted, without changing a line, to another journal. This really also depends on the reasons for rejection.

[EDIT] I agree that the peer review system requires confidentiality, and independence. Authors deserve several chances to get published.

[EDIT] Yet, some misconducts are becoming apparent. Redundant publications, plagiarism, for instance. This is not the case here. However, some authors do not hesitate to resubmit again and again, sometimes to journals with putative lower expectations, with hope they will finally go through, with some chance and unwary editors and reviewers.

Such practice floods the peer review system.

Since the paper already made it to the reviewers (and was already reviewed in the past), there are two options:

  1. let it flow without interference,
  2. interfere.

I am in favor, in the OP case, of the second option. I feel important to let the editor know about the situation, while performing the review. This would warn him from accepting the paper solely based on other lacky reviewers (which might be the intend of the submitting author: to get lucky with reviewers). I do feel an editor should be, at his place, capable of critical judgement on such a warning.

  1. What would be optimal is to have Journal 1 editor inform Journal 2 editor that he got aware of the situation, and inform Journal 2 editor about his decision for rejection. It is possible if you know well Journal 1 editor.
  2. Summmarize the main traits of the evaluation (including those from your co-reviewers for Journal 1), to save some time for your colleague, and invite him to write these concerns in the section "information to the editor only". I believe non-so-ethical to provide all the initial reviews to your colleague, yet, in extreme cases...
Laurent Duval
  • 4,995
  • 15
  • 35
  • 2
    I don't think it would be appropriate to tell the new editor that the paper had already been submitted-to and refereed at another journal. Nor to give the colleague a copy of the earlier referee report. – paul garrett Mar 08 '16 at 16:43
  • 1
    @paul garrett I fully aggree with the 2nd point. Could you please tell me more on why the 1st point does not fell appropriate when submitting exactly the same paper to another journal, if it is in the same domain? – Laurent Duval Mar 08 '16 at 16:50
  • 2
    Telling the editor that the paper had already been submitted may inadvertently prejudice them. That fact should be irrelevant to their judgement on it, which should be on its palpable merits-or-not alone. Any other action suggests collusion against the authors... even if not intended. – paul garrett Mar 08 '16 at 16:52
  • I understand the concern. Let me think about this to edit accordingly. The fact that both reviewers talked and compared the paper might prejudice the authors as well? This is not the case here, but cases of plagiarism, or redundant publication, are sometimes discovered with discussion between reviewers. – Laurent Duval Mar 08 '16 at 17:19
  • Possibly, this might depend on the field. In mathematics, I fully understand. I have concerns with this specific case with the somewhat "subjective" comment here that "originality of the paper was questionable, and content and presentation were severly lacking throughout" – Laurent Duval Mar 08 '16 at 17:22
  • 3
    You point about journals with lower expectations seems a bit off. Many journals have high expectations about "importance and impact", but I like to think that all respectable journals have high expectations about "correctness". – StrongBad Mar 08 '16 at 21:08
  • I am talking about all journals, including those one can pay to publish crap – Laurent Duval Mar 08 '16 at 21:26
  • 1
    I see no evidence supporting subjective interpretation. This is pretty black and white. The editor of the 2nd submission is not entitled to the history of the manuscript in this case. – Scott Seidman Mar 08 '16 at 22:27
  • 1
    Suggestion #1 (informing the editor) is terrible advice. You are effectively suggesting, that reviewer #2 admits to the journal, that he disclosed a confidential-for-review-only article to a 3rd person (reviewer #1). Unless reviewer #2 is feeling guilty for his misconduct, to the extent, that he's willing a deal a hit to his own reputation, but admit to wrongdoing, the best option is to decline to review and forget about it. – LLlAMnYP Mar 09 '16 at 17:16
  • 1
    @LLlAMnYP I shall discuss about such practice with additional colleagues in my field, to better fathom possible wrongdoing in my suggestion. – Laurent Duval Mar 09 '16 at 18:01